imagemetrics: Image Metrics

In gzt/monogamy: Monotone Spline Fitting with MGCV

Description

A dataset containing results of a survey asking for opinions about the quality of a set of distorted binary images. The data is in "long" format. There were 12 undistorted images ('base_img' and 'base_name') and 15 versions of each distorted image ('img_no', and 'distorted'). The 'mean' is the mean opinion score for each image: the average rating each respondent gave the distorted image compared to the base image. The standard deviation ('std') is also reported. Several different quantitative binary image quality metrics are also included, with their name in the 'metric' column and th value of the comparison in the 'score' column.

Usage

data(imagemetrics)

Format

An object of class tbl_df (inherits from tbl, data.frame) with 1120 rows and 8 columns.

Examples

head(imagemetrics)
aggregate(score ~ metric, imagemetrics, FUN = mean)
with(imagemetrics[imagemetrics$metric == "accuracy", ],
    plot(mean, score, xlab = "Accuracy", ylab = "Mean Opinion Score"))

gzt/monogamy documentation built on April 24, 2020, 9:45 p.m.